vaughn tan
@vt
Last week, I helped facilitate a meeting of AI industry with gov agencies thinking about the national implications of AI adoption. That 2-day conversation made me think that a strong focus on teaching specific tool use is maybe misguided. More here: https://uncertaintymindset.substack.com/p/49-the-work-of-uncertainty
1 reply
1 recast
0 reaction
Nick Drage
@pathdependence
Nice work, thank you for sharing. I'm not sure that I'd agree that exploitation of the worker is to the advantage of the employer though; maybe only in those standardised positions? Also, am I right in thinking that "negotiated joining" is your concept, and no-one else has picked it up in the last seven or so years?
2 replies
0 recast
1 reaction
vaughn tan
@vt
i gave "negotiated joining" a name and tried to clarify the concept, but aspects of the practice exist all over the place and have for at least centuries or millennia. (dating is an implementation of NJ tbh.) there are some companies and institutions, large and small, which have implemented NJ ...
1 reply
0 recast
1 reaction
Nick Drage
@pathdependence
I don't know about dating ( little personal experience ) but that area *feels* like it's not often NJ but just mismatched expectations. Do those companies and institutions understand NJ though... it feels like it might be something haphazard, whereas your overview could establish good practice or best practice?
1 reply
0 recast
1 reaction
vaughn tan
@vt
the basic idea of NJ is that you start out with some partly open-ended expectations about what you will do and what the counterparty will do — then you and the counterparty try stuff out, keeping what works and dropping what doesn't. if enough stuff sticks, y'all stay together.
1 reply
0 recast
1 reaction
vaughn tan
@vt
the basic idea is robust to misaligned expectations in that it still works but more slowly and irritatingly. so there's for sure ways to make the mutual discovery process more effective and/or speed it up — and i do talk about those, as well as scope conditions for applicability. the diagram below maybe clarifies?
2 replies
0 recast
1 reaction